Modeling stresses that develop in the heat treat process

[EngineerDave]

I am doing a side project that I haven't had time to research fully yet, but am posting in case someone can direct me to the correct starting research materials.

I want to model some thermal stresses that develop with regards to microstructural changes (martensitic transformations) in a part.

These thermal changes can lead to cracking and other problems.

In any event if anyone knows of some good case histories or references for thermal stress I'd love to read them.

I will use either ANSYS, IDEAS or NASTRAN (not sure which yet)

 

[toothroot]

Hi,
I had to perform a similar analysis two years ago under DE-FORM. ANSYS is also capable of that kind of analysis (multiphysics, coupling bewtween heat transfer and phase changes). I'm not sure if there's a NASTRAN package that can do phase changes; I-DEAS however cannot do such a highly nonlinear simulation (you might want to use it as pre/postprocessor, however).
After I spent some time analysing phase changes and the development of residual stresses, I must say, that this type of analysis is (given the current state of research and simulartion techniques) black magic. You will find that unless you have very good values for the material properties, your results are out of question. More or less all material data have to be defined with respect to temperature, carbon content and - in some cases - stress level. The phase transitions themselves are mathematically modelled by some kind of growth equations, with a number of coefficients. These you have to know as well.
After several attempts, I consider myself lucky to get results that are approximately what I would expect (which is, not totally out of order).
The following internet ressources might help you:

http://www.deform.com/

The FE-system DEFORM can perform highly nonlinear simulations and also phase transitions

http://www.esi-group.com/products/sysweld/index.php

Sysweld is another tool that can help you. The makers of Sysweld CLAIM they can analyse things such as welding or case-hardening.
I hope my post doesn't sound too discouraging, but you're entering the voodoo section of FEA. Let me know if you have any more specific question, maybe I can help you.

Cheers,
Daniel

 

[EngineerDave]

Thanks for the references.

It doesn't discourage me too much. It is more or less something I am taking on as a side project due to curiousity not a pressing project i need to work on.